I don't think you are straddling at all, instead you are just stating the obvious really. Too many folks have their heads up the butts dogmatically clinging to objective or subjective viewpoints. It's not a good / bad, black / white, objective / subjective world. The sooner they extract their heads and open their minds the better. You're spot on in my (sometimes-not-so-humble) opinion.
The issue is often a dogmatic one (as you can see by some of the responses here). A simple-minded approach looks at the power supply as a block and not as part of a system. I see audiophiles settle on The Answer and try to make that Answer be universal- and it never is.
Most passive filtering proponents must never had a good high voltage regulator in their systems, or maybe have used something badly excuted. All those considerations about bad tone with ss regs etc. are completely wrong in my practical experience. Plus passive is completely anti economic for $ and bulk.
P.S. Also for tube regs, a 6L6GC or alike CCS with a parallel stabilizer gas tube(s) OA-OD one is the better option IMHO.
P.S. Also for tube regs, a 6L6GC or alike CCS with a parallel stabilizer gas tube(s) OA-OD one is the better option IMHO.
Other possibilities:
They have an amp that is relatively immune to mediocre power supply regulation and feel that those big caps and chokes have to be doing something good. The small, efficient, and effective regulators seem like... cheating.
They like the coloration that a bouncing supply rail gives ("Wow, it's so much more dynamic!").
I'm sure there's more. "Bad regulation is a good thing" is not one of them.
There are two ways. One is emulating golden vintage tone, another is looking for performance. Since the hobby is about serving either our whim or can be headed to perfectionism, I accept both, but the confusion starts when we don't discriminate the order of priorities and goals. Strictly better is demonstrable by actual performance on the bench and higher information retrieval in audition. There are some shades of better considering subtle tone of linearity priorities also.
Well, I regulate the **** out of just about everything I design and build for myself. I currently have a stereo PSU under construction with 16 pairs of 200Wpdiss series pass transistors per channel for an SS amp that dissipates 1200W at idle, as well a a regulated PSU with a dozen series-pass 12E1's, for a valve amp.
I also have commercial Williamson type amp with a passive supply that would out perform almost anything I've ever seen published here by an order of magnitude or more. It has a SS rectifier supply and I've bunged a big fat electro in it, and that gives me a 20Hz supply impedance that would rival that of a primitive tube shunt regulator. Would a regulated supply with an even lower LF output impedance improve the sonic qualities of this amp? Nope, not in my opinion. There is some rather low-level residual hum though, that could be banished, but this is only obvious if you stick your head up to a speaker.
And yeah, I sure can here the difference between this one and a primitive SET amp (which no amount of regulation could bring to the same level).
Last edited:
Having a nice tube amp with good psrr that performs with just a capacitive filter isn't the norm. I suppose we talk circuits that need better attention. Your example proves the exemption that proves the rule.
This is late 1950's technology!
I don't think you are straddling at all, instead you are just stating the obvious really. Too many folks have their heads up the butts dogmatically clinging to objective or subjective viewpoints. It's not a good / bad, black / white, objective / subjective world. The sooner they extract their heads and open their minds the better. You're spot on in my (sometimes-not-so-humble) opinion.
My only question is "What are you comparing your final designs to?" If you are not comparing your designs against a known "reference" caliber commercial product, how do you detrermine the true nature of what you have created? Does it sound better than the reference? Does it sound more real, with proper soundstaging and transparency? If there is no reference, how can you determine if the design is a success or a failure by measuring alone?
CW; perhaps you misunderstood, but no one is suggesting 'by measuring alone'.
Hi Salas
I am dealing with this at present. My prototype preamp has served as a PS test bed over the last few months and here is what i've tried:
- CLCRC with 8H 36ohm Lundahl choke and Cerafines
- CRCRC
- series reg shamelessly lifted off ARC REF3
- CCS (DN2540) feeding an OD3
- CCS feeding Salas shunt reg
Rectification is from a 5Y3GT and i flatly refuse to consider SS rectification.
Out of all the schemes the best compromise comes from the "Salas" shunt reg but with either a CLC or LC filter in front. This is quite madenning as although a CRCRC filter provides much lower ripple than the LC, the sound is much worse. I also have a large number of old chokes, many by Partridge, but none provide the air and dynamics of the Lundahl. So, i guess no free lunch for me
Out of all the schemes the best compromise comes from the "Salas" shunt reg but with either a CLC or LC filter in front. This is quite madenning as although a CRCRC filter provides much lower ripple than the LC, the sound is much worse. I also have a large number of old chokes, many by Partridge, but none provide the air and dynamics of the Lundahl. So, i guess no free lunch for me
You mean the ''Simplistic HV Shunt Reg''? If yes, there is a ''prefiltered'' recommendation there in its thread, because I have developed it with a CLC infront of it right from the start. Good it works nicely for you too. Many people like it a lot. Its not a high performance circuit so the less DC resistance infront of it given by an inductor VS CRCRC probably swings less nasties there. It has better ripple rejection than pristine output impedance as a reg, you see.
Instead of being so dogmatic, how about working out the technical side of why they do not perform well, and then look for alternatives?
Did the guy at the back say "Schottky" ?
There's an Italian fellow, whose name I don't remember, who makes amps that look like someone dumped the whole Lundahl catalog onto a chassis. There seem to be many schools of fashion- many, many amps that are designed rather than engineered.
If you haven't seen big caps and chokes, you haven't been looking at many DIY projects.
If you haven't seen big caps and chokes, you haven't been looking at many DIY projects.
I have been concerned about “starving” the pentode with <1mA plate current.
The typical data for the 6sj7 is 3mA @ plate, G2 100V @ .85mA.
I will test this tonight… I have some 10M45’s to use for a ccs.
One question: If I use a ccs on the pentode plate, How do I adjust the B+ voltage ?
I would suspect the ccs to hold the pentode plate at the same voltage regardless of how I adjust the G1 voltage?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.